About me

Hi, I'm Aryan Dhir—a robotics enthusiast and programmer with a strong interest in building intelligent systems that solve practical problems. I enjoy working at the intersection of hardware and software, where code brings machines to life and makes them useful in the real world.

Over time, I’ve worked on a variety of robotics projects, each one helping me develop my skills in electronics, control systems, and embedded programming. One of my main projects is a remote-controlled Bluetooth hovercraft designed to clean ponds and lakes, aimed at addressing small-scale water pollution. It’s fast, maneuverable, and built to handle real-world outdoor conditions.

I also built a line-following robot that uses an OpenMV camera to detect and classify multi-colored balls, and then sort them accordingly. This project introduced me to basic computer vision and object recognition, and it demonstrated how robotics and machine learning can be combined to automate tasks more intelligently.

On the more experimental and fun side, I’ve created a remote-controlled omni-directional car, which can move in any direction using mecanum wheels, as well as a custom-built speed boat that I designed from scratch for performance and control testing. These projects helped me better understand motor control, power systems, and wireless communication.

Outside of robotics, I’ve developed a strong interest in machine learning and its applications in signal processing. One of my current projects involves removing stutters and disfluencies from recorded speech using deep learning models. The goal is to take real-world audio, analyze the patterns of hesitation or repetition, and generate smoother, cleaner output without losing the speaker’s natural tone.

There are many practical applications for this kind of technology—from improving recorded content like podcasts or voiceovers to assisting individuals with speech disorders in producing more fluent audio. It’s a challenging problem that involves natural language processing, audio segmentation, and time-based signal reconstruction, and I’m learning a lot through the process.

I created Mind of Machines to showcase my work, document my progress, and share what I learn along the way. I believe in building things that are technically interesting and practically useful—whether that means automating a task, solving an environmental problem, or experimenting with machine learning in new ways.

Right now, I’m continuing to expand my skills in robotics, AI, and systems integration. I’m always looking for new ideas, tougher problems, and better tools to build with. If it moves, thinks, or learns—I want to understand how and why.

Thanks for checking out my work.